Am.  Jour.  Pharm.  \ 
December,  1899.  J 
Analysis  of  Essential  Oils. 
579 
use  of  a  capillary  tube  is  indispensable,  to  prevent  bumping,  the 
form  Fig.  6  of  flasks  is  very  convenient ;  the  tube  /  takes  up  the 
capillary.    These  flasks  are  an  article  of  commerce. 
To  avoid  overheating  of  the  liquid  and  to  secure  a  very  even  dis- 
tillation, the  use  of  a  bath  instead  of  the  heating  with  free  flame  is 
recommended.  A  bath  of  any  fatty  oil,  or  of  liquid  paraffine,  will 
generally  be  serviceable;  more  convenient,  however,  because  of  not 
.being  inflammable  and  not  giving  off  any  offensive  vapors  at  high 
temperatures,  is  a  bath  from  Wood's  metal,  which  melts  at  about 
6o°  C;  it  is  true  that  the  first  cost  is  somewhat  high,  but  it  will 
last  a  lifetime. 
FIG.  4 
A  very  common  fault  of  beginners  is,  to  take  Irom  the  start 
fractions  within  very  narrow  intervals  of  temperature,  in  the 
opinion  thus  to  effect  a  quicker  fractionation.  Soon  they  then 
have  such  a  large  number  of  small  fractions  that  the  indications  of 
the  thermometer  become  quite  uncertain  because  the  rise  in  dis- 
tillation temperature  of  the  successive  fraction  is  so  fast  that  an 
exact  reading  off  is  impossible.  Such  a  proceeding  also  appears 
to  be  quite  misleading,  if  we  consider  the  very  imperfect  sepa- 
ration during  the  first  fractionations.  It  is  much  better  first  to 
divide  the  whole  oil  into  four  or  five  about  equal  fractions,  without 
